Graphene oxide derivatives as hole- and electron-extraction layers for high-performance polymersolar cells
Jun Liu,*a Michael Durstockb and Liming Dai*c
Owing to their solution processability, unique two-dimensional structure, and functionalization-induced
tunable electronic structures, graphene oxide (GO) and its derivatives have been used as a new class of
efficient hole- and electron-extraction materials in polymer solar cells (PSCs). Highly efficient and stable
PSCs have been fabricated with GO and its derivatives as hole- and/or electron-extraction layers. In this
review, we summarize recent progress in this emerging research field. We also present some rational
concepts for the design and development of the GO-based hole- or electron-extraction layers for high-
performance PSCs, along with challenges and perspectives.
Broader context
Owing to their solution processability, unique two-dimensional structure, and functionalization-induced tunable electronic structures, graphene oxide (GO) andits derivatives have been used as a new class of efficient hole- and electron-extraction materials in polymer solar cells (PSCs). Highly efficient and stable PSCshave been fabricated with GO and its derivatives as hole- and/or electron-extraction layers. In this review, we summarize recent progress in this emergingresearch eld. We also present some rational concepts for the design and development of the GO-based hole- or electron-extraction layers for high-performancePSCs, along with challenges and perspectives.
1. Introduction
During the past decade or so, polymer solar cells (PSCs) haveattracted a great deal of interest because of many competitiveadvantages, including their versatility for large-scale fabricationthrough the roll-to-roll process, exibility, lightweight, and lowcost.1–9 Continued development of novel polymeric/organicmaterials, optimization of device structures, and improvementof fabrication techniques have steadily increased the powerconversion efficiency (PCE) of PSCs up to >9% for single junc-tion cells10 and >10% for tandem cells.11 In order for PSCs to becompetitive with conventional photovoltaic technologies basedon silicon or other inorganic materials, however, the efficiencyand lifetime of PSCs still need to be signicantly improved.
The photovoltaic effect involves generation of electrons andholes in a semiconductor device under illumination, andsubsequent charge collection at opposite electrodes. Photonabsorption of organic optoelectronic materials oen creates
bound electron–hole pairs (i.e. excitons). Charge collection,therefore, requires dissociation of the excitons, which occursonly at the heterojunction interface between semiconductingmaterials of different ionization potentials or electron affinities.Like many other polymeric thin lm devices,12 the interfaces inPSCs play critical roles in regulating the charge separation andcharge collection, and hence the overall device performance.13,14
For high-performance PSCs, the work functions of the cathodeand the anode need to match the LUMO level of the acceptorand the HOMO level of the donor, respectively, to minimizeenergy barriers for electron- and hole-extraction. The energybarriers between the active layer and the electrodes can also beeffectively reduced by electron-/hole-extraction layers at thecathode/anode. Therefore, a hole-extraction layer (HEL)between the anode and the active layer, as well as an electron-extraction layer (EEL) between the cathode and the active layer,are essential for achieving maximum PSC device efficiency andlifetime.15–17 The functions of hole- and electron-extractionlayers include: (i) to minimize the energy barrier for chargeextraction; (ii) to selectively extract one sort of charge carrierand block the opposite charge carrier; (iii) to improve theinterface stability between the electrode and the active layer; (iv)to modify the surface properties and alter the active layermorphology; and (iv) to act as an optical spacer.15–17
Several classes of materials, including organic conductivepolymers (e.g. poly(3,4-ethylenedioxythiophene) doped with
poly(styrenesulfonate), PEDOT:PSS), self-assembled mono-layers (e.g. 3,3,3-triuoropropyltrichlorosilane) and metal oxideinorganic semiconductors (e.g. NiO, MoO3, V2O5, WO3), havebeen used as HELs.15–17 The most widely used HEL in PSCs isPEDOT:PSS. However, PEDOT:PSS suffers from its strong acidity(pH ¼ 1–2) and hygroscopicity, etching the indium tin oxide(ITO) electrode to cause degradation of the device efficiency andlifetime.18,19 On the other hand, the metal oxide inorganicsemiconductors oen need to be thermally deposited underhigh vacuum and are incompatible with the high throughoutroll-to-roll process of PSCs. Therefore, recent efforts have beendevoted to developing solution-processable metal oxide semi-conductors by the sol–gel approach or colloid nanoparticleapproach.20–22 The materials used as EELs include low workfunction metals or related salts (e.g. Ca, LiF), metal oxidesemiconductors (e.g. TiO2, ZnO), fullerene derivatives, andconjugated polyelectrolytes.15–17 The conjugated polyelectrolyteEELs have recently given a record high PSC device efficiency10
while metal oxide semiconductor EELs have been widely usedwith the additional advantages of resistance to oxygen andmoisture as well as optical transparency.
Owing to its extraordinary mechanical, electrical, optical,and thermal properties, the two-dimensional (2D) single-atomic-thick sp2-hybridized carbon sheet of graphene hasquickly emerged as an attractive candidate for energy applica-tions.23–26 However, graphene sheets without functionalizationare insoluble and infusible with limited practical applications.Recent efforts have led to solution-processable graphene oxides(GOs) from exfoliation of graphite powders with strongoxidizing reagents (e.g. HNO3, KMnO4 and/or H2SO4).27,28 Theavailability of reactive carboxylic acid groups at the edge andepoxy/hydroxyl groups on the basal plane of GO sheets facili-tates functionalization of graphene, allowing tunability of opto-electronic properties while retaining the good solubility in wateror polar organic solvents.29,30 Moreover, GO can be producedand processed in solution at large scale with low cost, particu-larly attractive for massive applications. Indeed, GO and itsderivatives have been demonstrated to be useful in manyapplications with excellent performance, such as batteries,supercapacitors, fuel cells, solar cells, sensors, catalysts andcomposite materials.31–33 Of particular interest, GO materialshave been used in every part of PSC devices, including as elec-trodes, charge extraction layers, and in the active layer.34–37 Inthis review, we summarize the development of GO materials ascharge extraction layers in PSCs by presenting some rationalconcepts for the design and development of the GO-based hole-or electron-extraction layer for high-performance PSCs, alongwith the challenges and perspectives in this emerging researcheld.
2. Graphene oxide derivatives as thehole extraction layer
In 2010, Li et al.38 reported the rst use of GO as an efficient HELin PSCs because of its suitable work function and good lm-forming property. This work triggered the recent extensive
1298 | Energy Environ. Sci., 2014, 7, 1297–1306
research on GO as charge extraction layers in PSCs. Comparedwith other charge extraction materials, GO possesses manyunique advantages, including its two-dimensional structure,easy functionalization, tunable energy levels, solution process-ability and low cost. With the various strategies developed forimproving GO performance in PSCs, several highly efficient andstable PSCs have been reported with GO as the HEL.
To facilitate the hole collection and extraction at the anode,the hole extraction material should have a proper work functionto ensure an Ohmic contact with the donor material for efficienthole transport without increasing the device series resistance.As demonstrated by Li et al.,38 GO had a work functionof�4.7 eV tomatch well with the poly(3-hexylthiophene) (P3HT)donor material for efficient hole extraction (Fig. 1b). Besides,GO could be uniformly deposited onto an ITO anode (Fig. 1b)simply by spincoating its aqueous solution. The PSC device withP3HT:[6,6]-phenyl-C61-butyric acid methyl ester (PCBM) as theactive layer and GO as the HEL (see Fig. 1a) exhibited an open-circuit voltage (VOC) of 0.57 V, short-circuit current density ( JSC)of 11.40 mA cm�2, ll factor (FF) of 0.54, and power conversionefficiency (PCE) of 3.5� 0.3%. This value of PCE is much higherthan that (PCE ¼ 1.8%) of the device without HEL and is fairlycomparable to that (PCE ¼ 3.6%) of the device based on thestate-of-the-art HEL, PEDOT:PSS (Fig. 1d). With the GO layerthickness increased from 2 nm to 10 nm, the FF of the devicedramatically decreased from 0.54 to 0.19 with a concomitantdecrease in PCE from 3.5% to 0.9% (Fig. 1e) due to theincreased series resistance with increasing thickness arisingfrom the insulating nature of GO. The insulating property of GOand its thickness-dependent performance in PSC devices aredisadvantages for GO HEL, which will be addressed later inmore detail with possible solutions.
Gao et al.39 reported the utilization of GO as the HEL ininverted PSCs. The device conguration is shown in Fig. 2a. Theuniform GO layer (Fig. 2b) was deposited on the P3HT:PCBMactive layer by spincoating its solution in anhydrous butylalcohol. The inverted PSC device with optimal GO layer thick-ness (2–3 nm) showed a VOC of 0.64 V, JSC of 8.78 mA cm�2, FF of0.64, and PCE of 3.60%. This performance is also fairlycomparable to that of the PEDOT:PSS-based inverted device(Fig. 2c). These authors further discovered that GO could dopeP3HT at the surface of the active layer. This was because GOcontains carboxylic groups, phenolic and enolic groups withhigh content of protons.40 The heavily doped P3HT thin layer atthe interface facilitated the formation of an Ohmic contactbetween the active layer and the top metal electrode, and henceleads to much enhanced device performance.
As mentioned above, one drawback of the GO HEL is itsinsulating nature, leading to an increased series resistance witha concomitant decrease in FF and PCE of the resulting device.As shown in Fig. 1, those epoxy and hydroxyl groups on thebasal plane of GO disrupt the sp2 conjugation of the graphenelattice to make GO an insulator.41 As a result, PSCs based on aP3HT:PCBM active layer and GO HEL always exhibit a FF lessthan 0.65 while the typical value for high performancePEDOT:PSS-based devices is about 0.70. Besides, the deviceperformance is highly sensitive to the thickness of the GO layer
Fig. 1 (a) Schematic illustration of the PSC device structure with GO as the HEL. (b) Energy level diagrams of the bottom electrode ITO, interlayermaterials (PEDOT:PSS, GO), P3HT (donor), and PCBM (acceptor), and the top electrode Al. (c) An AFM height image of a GO thin film with athickness of approximately 2 nm. (d) Current density–voltage (J–V) characteristics of the devices with no HEL, with 30 nm PEDOT:PSS film, andwith 2 nmGO film. (e) J–V characteristics of the ITO/GO/P3HT:PCBM/Al devices with the GO layer of different thicknesses. Adapted from ref. 38with permission from the American Chemical Society.
Fig. 2 (a) Device configuration of the inverted PSCs using the GOinterfacial layer (IFL) as the HEL. (b) AFM height image of the GO layeron glass/ITO/ZnO/C60-SAM/P3HT:PCBM stacks. (c) J–V characteris-tics of the inverted PSCs without an interface layer, with a 50 nmPEDOT:PSS layer, and with a 2.1 nmGO layer. Reproduced from ref. 39with permission from American Institute of Physics.
Review Energy & Environmental Science
Publ
ishe
d on
20
Nov
embe
r 20
13. D
ownl
oade
d by
CA
SE W
EST
ER
N R
ESE
RV
E U
NIV
ER
SIT
Y o
n 28
/04/
2014
02:
44:0
8.
View Article Online
(see Fig. 1e).38 To construct high-performance PSC devices withGO as the HEL, therefore, the conductivity of the GO layer mustbe signicantly improved.
The post-oxidation reduction to remove the oxygen-con-taining groups for recovering the conjugated structure of thebasal plane has been utilized to improve the GO conductivity.42
It has been found that GO could be reduced into r-GO by variousapproaches, including thermal annealing, microwave irradia-tion, laser irradiation, and chemical reduction in solution. Alarge variety of reagents, such as hydrazine, NaBH4, vitamin C,KOH, and HI, have been used to chemically reduce GO insolution.43 The solution-reduction of GO into reduced grapheneoxide (r-GO) is highly compatible with the solution-basedfabrication of PSCs. However, r-GO with a reduced number ofoxygen-containing groups oen exhibits poor solubility incommon solvents and tends to form aggregates in the disper-sion. Hence, r-GO cannot afford a uniform thin lm depositionby spincoating. The poor solubility of r-GO prevents its appli-cation as the solution-processed HEL in PSCs. Severalapproaches, including development of soluble reduced gra-phene oxide with specic reduction reagents,44–46 post-treat-ment of GO lm for reduction aer spincoating GO aqueoussolution,47,48 and introducing some highly conductive ller tothe GO thin lm,49 have been demonstrated to effectivelycircumvent the poor solubility of r-GO.
Yun et al.44 developed a solution-processable reduced gra-phene oxide (pr-GO) through reducing GO with p-toluene-sulfonyl hydrazide in an aqueous solution. pr-GO wasdemonstrated to be an excellent HEL for efficient and stablePSCs. For comparison, they also prepared normal r-GO byreducing GO with hydrazine as widely used in the literature.28
Both pr-GO and r-GO showed about 105 times higher conduc-tivity than that of GO. Like GO, pr-GO could also be uniformlysolution-cast on the ITO surface. In contrast, r-GO could notafford a uniform lm deposition and formed large aggregates
Fig. 3 AFM height images of GO (a), r-GO (b), and pr-GO (c) spincoated on the ITO electrode. (d) J–V characteristics of the PSC devices with GO,r-GO, and pr-GO as the HEL. (e) Changes in PCE of a conventional PEDOT:PSS-based PSC device and a pr-GO-based PSC device duringexposure to air. Reproduced from ref. 44 with permission from John Wiley and Sons.
Fig. 4 (a) Schematic structure and synthetic route to GO-OSO3H. (b) J–V curves of the PSC devices with PEDOT:PSS (25 nm), GO (2 nm), or GO-OSO3H (2 nm) as the HEL. (c) J–V curves of the PSC devices with GO-OSO3H as the HEL with different thicknesses. Reproduced from ref. 45 withpermission from the American Chemical Society.
Energy & Environmental Science Review
Publ
ishe
d on
20
Nov
embe
r 20
13. D
ownl
oade
d by
CA
SE W
EST
ER
N R
ESE
RV
E U
NIV
ER
SIT
Y o
n 28
/04/
2014
02:
44:0
8.
View Article Online
on the ITO surface (Fig. 3a–c). Consequently, pr-GO exhibitedmuch better PSC device performance than those of GO and r-GO(Fig. 3d). The PSC device with P3HT:PCBM active layer and pr-GO as the HEL showed a PCE of 3.63% with the FF of 0.667, JSC
1300 | Energy Environ. Sci., 2014, 7, 1297–1306
of 9.33 mA cm�2, and VOC of 0.59 V, which was highly compa-rable to its PEDOT:PSS counterpart. Moreover, the pr-GO-baseddevice manifested much longer lifetime than that of thePEDOT:PSS-based device (Fig. 3e).
Liu et al.45 reported a rationally designed solution process-able sulfated graphene oxide (GO-OSO3H) synthesized bytreating GO with fuming sulfuric acid to introduce –OSO3Hgroups onto the reduced basal plane of GO (Fig. 4a). Theseauthors found that the dehydration effect of fuming sulfuricacid resulted in reduction of the carbon basal plane andenhanced conductivity (1.3 S m�1 for GO-OSO3H vs. 0.004 S m�1
for GO). Furthermore, the newly introduced –OSO3H groups,along with the existing –COOH groups in GO, rendered GO-OSO3H soluble (1 mg mL�1 DMF solution) for solution pro-cessing. In addition, the protons in –OSO3H groups and –COOHgroups in GO-OSO3H gave rise to surface doping of P3HT in theactive layer.40 As a result, PSC devices with GO-OSO3H as theHEL showed excellent device performance with an extraordi-narily high FF of 0.71, VOC of 0.61 V, JSC of 10.15 mA cm�2, andPCE as high as 4.37% (Fig. 4b). This performance is among thehighest reported for PSC devices with the P3HT:PCBM activelayer. Moreover, the device performance was nearly indepen-dent of the GO-OSO3H layer thickness (Fig. 4c), which was instark contrast to the aforementioned device with the insulatingGO as the HEL.
Apart from the aforementioned approaches, post-thermalannealing of the preformed GO lm has also been demonstratedto obtain reduced GO thin lm with enhanced conductivity.During thermal annealing, oxygenated functional groups in GOcan be removed via release of gas molecules (e.g. H2O, CO2, CO)
Fig. 5 XPS spectra of GO (a) without thermal treatment and (b) withthermal treatment at 250 �C for 10 minutes. (c) J–V characteristics ofPSC devices using thermally reduced GO as the HEL. Reproducedfrom ref. 47 with permission from Elsevier.
from the reduced carbon basal plane.43However, this approach islimited to PSC devices on glass substrates. Plastic substrates inexible PSCs cannot tolerate the high temperature (200–300 �C)required for the thermal GO reduction.
Joen et al.47 demonstrated efficient PSCs with thermallyannealed GO layer as the HEL. The GO HEL was deposited byspincoating GO aqueous solution on the glass/ITO substrate,followed by thermal annealing at 150 �C, 250 �C or 350 �C for10 minutes in air. The reduction of GO was conrmed by thedramatic decrease of the C–O peak intensity in the X-rayphotoelectron (XPS) spectra (see Fig. 5a and b). Take thethermal annealing at 250 �C as an example, the GO lmconductivity increased from 8 � 10�6 S m�1 to 1.8 S m�1 aerthe reduction by thermal annealing. Consequently, the PCE ofthe PSC device increased from 1.47% to 3.98%, which was fairlycomparable to that of the control device with PEDOT:PSS as theHEL. Moreover, the device based on thermally annealed GOexhibited much better stability than the PEDOT:PSS-baseddevice. Liu et al.48 had systematically investigated the inuenceof the preparation conditions of GO HEL on the PSC deviceperformance. They also found that high temperature (230 �C)treatment of GO HEL aer spincoating increased the GO layerconductivity and greatly improved the FF and PCE of theresulting PSC devices. To make high quality GO lms, theseauthors optimized the concentration and spincoating speed ofthe GO solution.
In addition to the thermal treatment, other treatments ofpreformed GO thin lms have also been investigated to improvethe performance of devices with GO as the HEL. For instance,Murray et al.50 reported a highly efficient and stable PSC withGO as the HEL and PTB7:PC71BM (Fig. 6a) as the active layer. Inthis case, the GO layer was deposited onto a clean ITO substrate
Fig. 6 (a) Chemical structures of the PTB7 donor and PC71BMacceptor. (b) J–V plots under AM 1.5 G illumination for PSCs withPEDOT:PSS and GO as the HEL. (c) Thermal degradation of encap-sulated devices at 80 �C under an N2 atmosphere. (d) Environmentaldegradation of unencapsulated devices fabricated with air-stableelectrodes at 25 �C under 80% relative humidity. Reproduced from ref.50 with permission from the American Chemical Society.
with a controlled density via Langmuir–Blodgett assembly, fol-lowed by low level ozone exposure to modify the GO surfacechemistry. These authors found that the GO HEL thus preparedcould effectively regulate the PTB7 p-stacking orientation to befavorable for charge extraction. The resultant GO-based deviceshowed a PCE of 7.39%, comparable favorably to the corre-sponding value of 7.46% for a PEDOT:PSS-based device(Fig. 6b). More importantly, this GO-based device provided a5 times enhancement in thermal aging lifetime (Fig. 6c) and a20 times enhancement in humid ambient lifetime (Fig. 6d) withrespect to the PEDOT:PSS-based device. Yang et al.51 treated GOHEL with oxygen plasma to change the surface characteristicsfor an improved PSC efficiency. It was found that the oxygenplasma treatment led to simultaneous enhancements in JSC(9.91 mA cm�2 vs. 8.42 mA cm�2), VOC (0.60 V vs. 0.58 V), and FF(0.60 vs. 0.55), and hence a signicantly improved PCE (3.59%vs. 2.71%), which was attributed to a high work function of theoxygen plasma treated GO for ensuring an increased holemobility and enhanced hole extraction.
The potential use of conductive llers to improve theconductivity of GO HELs has also been investigated. In thiscontext, single-walled carbon nanotubes (SWCNTs) with a smalldiameter of ca. 1 nm are particularly attractive because they donot signicantly increase the surface roughness of the GO thinlm. However, controlling the quality of dispersion to obtainhomogenous conductively lled GO lms is critical for the
Fig. 7 (a) Addition of a small amount of SWCNTs into the GO film candecrease the through-thickness resistance of the GO film by an orderof magnitude. (b) Addition of a small amount of SWCNTs into the GOlayer can increase the FF and JSC of devices with GO HEL. Reproducedfrom ref. 49 with permission from John Wiley and Sons.
1302 | Energy Environ. Sci., 2014, 7, 1297–1306
performance of the resultant PSC devices. Kim et al.49 havesuccessfully incorporated SWCNTs into the GO HEL by mixingSWCNTs and GO in water under sonication, followed by spin-coating to yield a uniform GO:SWCNT composite thin lm. Ascan be seen in Fig. 7a, the incorporation of a small amount ofSWCNTs increased the through-thickness conductivity of theGO lm by an order of magnitude. Therefore, the incorporationof SWCNTs into the GO layer not only improved the deviceefficiency, but also allowed the use of thicker and easier-to-make GO lms. PSC devices with P3HT:PCBM active layer andGO:SWCNT ¼ 1:0.2 as the HEL exhibited a VOC of 0.60 V, JSC of10.82 � 0.56 mA cm�2, FF of 0.628 � 0.0031, and PCE of 4.10 �0.18%, which were fairly comparable to those of the PEDOT:PSS-based device (Fig. 7b). Having demonstrated the excellent deviceperformance for GO:SWCNT as the HEL, the same authorsfurther usedGO:SWCNT as an interconnect layer for constructingserially connected tandem polymer solar cells with subcellsstacked along the optical path to increase optical absorption.52
The key index of a successful tandem structure is the value of VOC,which ideally should be the sum of VOC of the constituent sub-cells. Both regular and inverted tandem cells with GO:SWCNT asthe interconnect layer were constructed to show a VOC of 84% and80% of the sum of the two constituent subcells, respectively.These results indicate that successful serial connection of sub-cells with the GO:SWCNT has been achieved.
In addition to the use of SWCNTs as the conductive ller,PEDOT:PSS has also been blended into theGOHEL for improvingthe PSC performance.53–56 As demonstrated by Tung et al.,55 themixing of GO and PEDOT:PSS in water caused the dispersion toincrease its viscosity dramatically, producing a sticky lm uponsolution casting, due to possible PEDOT chain reorientationaround GO sheets. Therefore, tandem PSCs could be fabricatedby a direct adhesive lamination process with the sticky conductiveGO:PEDOT lm. Alternatively, Fan et al.54 have introduced gra-phene oxide decorated with Au-nanoparticles (Au NPs) into aPEDOT:PSS layer for utilizing the plasmonic effect associatedwith the Au NPs to increase light absorption and improve the JSCand PCE of the PSC device. The decoration of Au NPs on GOsheets could ensure a uniform dispersion of Au NPs, whichotherwise could easily aggregate in a physically blended system.
The use of a GO and metal oxide bilayer HEL has also beendemonstrated to show excellent PSC device performance.Indeed, Ryu and Jang56 reported that PSCs with HELs based onGO, NiOx and a GO/NiOx bilayer exhibited PCEs of 2.33%, 3.10%,and 3.48%, respectively. The bilayer structure HEL showed thehighest PCE with the VOC of 0.602 V, JSC of 8.71 mA cm�2, and FFof 0.664. Furthermore, Chao et al.57 have successfully used a GO/VOx bilayer as the HEL to construct high-performance invertedPSCs (Fig. 8a). Inverted PSC devices with solution-processedmetal oxide HEL oen suffer from a relatively low VOC because ofthe penetration of metal oxide precursors into the underlyingactive layer to form defect sites during the lm formation byspincoating. To avoid the interfacial penetration, they placed athin layer of two-dimensional graphene oxide sheet between theactive layer and the metal oxide layer. The high LUMO level of theGO interlayer could also block electrons to promote VOC (Fig. 8a).As a result, the P3HT:PCBM-based inverted device with a GO/VOx
Fig. 8 (a) Schematic illustrations of the device architecture and theenergy level diagram of solution-processed inverted PSCs with GO/VOx bilayer as the HEL. (b) J–V characteristics of P3HT:PCBM invertedPSCs with different HELs. Reproduced from ref. 57 with permissionfrom John Wiley and Sons.
Review Energy & Environmental Science
Publ
ishe
d on
20
Nov
embe
r 20
13. D
ownl
oade
d by
CA
SE W
EST
ER
N R
ESE
RV
E U
NIV
ER
SIT
Y o
n 28
/04/
2014
02:
44:0
8.
View Article Online
bilayer HEL showed a PCE of 4.1%, outperformed its counter-parts with only GO or VOx as the HEL (Fig. 8b). It was worthy tonote that the GO/metal oxide bilayer concept has also beenapplied to high-efficiency PSCs based on low bandgap polymerdonors. An inverted device with a low bandgap polymer
Fig. 9 (a) Schematic illustration of synthesizing graphene oxide ribbon (the GOR-based PSC device. (c) J–V curves under AM 1.5 G illuminatioReproduced from ref. 60 with permission from John Wiley and Sons.
(PTh4FBT) as the donor material and bilayer GO/VOx as the HELexhibited a PCE of 6.7%, which was fairly comparable to 6.8% ofthe corresponding control device with vacuum-evaporated MoO3
as the HEL. Besides, Park et al.58,59 investigated the bilayerstructure of GO or r-GO/PEDOT:PSS as the HEL in PSCs, andfound that the GO/PEDOT:PSS bilayer showed a better deviceperformance than that of the r-GO/PEDOT:PSS bilayer. This waspresumably due to the different work functions of r-GO (fF ¼ 4.5eV) and GO (fF ¼ 4.7 eV).
Along with the extensive research on the development of GOderivatives as HELs for high-performance PSCs, Liu et al.60 haverecently developed graphene oxide ribbon (GOR) as the HEL.GOR was prepared by oxidative unzipping of SWCNTs, followedby an extra oxidation process (Fig. 9a). The GOR combined thesolution processability of GO and the semiconducting propertywith a bandgap of graphene ribbon. Cyclic voltammetry of GORimplied a HOMO of�5.0 eV and LUMO of �3.5 eV, which couldfacilitate hole transporting and electron blocking to minimizethe electron–hole recombination on the anode (Fig. 9b). More-over, GOR showed superior lm forming properties on the ITOsurface. Consequently, PSCs based on GOR HEL exhibited a VOCof 0.62 V, JSC of 9.96 mA cm�2, FF of 0.67, and PCE of 4.14%.This performance was comparable to that of the PEDOT:PSS-based device and was much better than that of the corre-sponding GO-based device (Fig. 9c). Furthermore, the PSCdevices with GOR as the HEL exhibited a better stabilitycompared to the PEDOT:PSS-based device.
3. Graphene oxide derivatives as theelectron extraction layer
While HEL materials should have a high work function, elec-tron extraction layer (EEL) materials should have a low work
GOR) by oxidative unzipping of SWCNTs. (b) Energy level alignment ofn of the PSCs without and with GO, PEDOT:PSS or GOR as the HEL.
Fig. 10 (a) Schematic structure and synthetic route to GO-Cs. (b)Energy level diagram of the PSC device with GO as the HEL and GO-Csas the EEL. (c) J–V curves of devices (device structure: ITO/PEDOT:PSS/P3HT:PCBM/EEL/Al) with none, LiF, Cs2CO3, and GO-Csas the EEL. Reproduced from ref. 62 with permission from John Wileyand Sons.
Fig. 11 Schematic illustration of the fabrication steps of BHJ solar cellswith GO as the EEL by stamping transfer. (a) Attachment of the transferfilm on top of the active layer; (b) after detachment of the film, the firstEEL of GO was uniformly transferred and coated onto the active layer;(c) spin-casting TiOx as the second EEL on top of GO; (d) completeddevice structure after Al deposition. Energy-level diagrams of theactive layer with EELs of TiOx (e) and GO/TiOx (f). Evac ¼ vacuum level,EF ¼ Fermi level, D ¼ interfacial dipole, fh ¼ hole-injection barrier. (g)J–V characteristics of devices without EEL and with the EEL of TiOx,GO, and GO/TiOx. Reproduced from ref. 64 with permission fromJohn Wiley.
Energy & Environmental Science Review
Publ
ishe
d on
20
Nov
embe
r 20
13. D
ownl
oade
d by
CA
SE W
EST
ER
N R
ESE
RV
E U
NIV
ER
SIT
Y o
n 28
/04/
2014
02:
44:0
8.
View Article Online
function to match the LUMO level of the acceptor material inthe active layer to facilitate electron extraction. Moreover, EELsshould efficiently transport electrons to minimize series resis-tance of the PSC devices for good photovoltaic performance.Being an ambipolar material for efficient transport of bothholes and electrons,61 GO derivatives with tunable energy levelsby functionalization can also be used as the EEL for PSCs. Thus,GO and its derivatives are the rst class of charge extractionmaterials which can be used as both HELs and EELs.
Liu et al.62 reported the rst GO-based electron extractionmaterial by using the cesium-neutralized graphene oxide (GO-Cs) as the EEL for PSCs. As schematically shown in Fig. 10a,upon adding Cs2CO3 into an aqueous solution of GO, theperiphery –COOH groups on the graphene oxide sheets wereneutralized and afforded –COOCs groups in the resultantGO-Cs. GO-Cs modied electrode showed the work function of4.0 eV, matching well with the LUMO level of the PCBMacceptor. Thus, GO-Cs could be used as the EEL in PSCs. In fact,PSC devices with GO-Cs as the EEL and P3HT:PCBM active layerexhibited fairly comparable VOC, JSC, FF and PCE with those ofthe corresponding control device with the state-of-the-art EEL(e.g. LiF), indicating that GO-Cs was indeed an excellent EEL.These authors further fabricated regular and inverted PSCs withboth GO as the HEL and GO-Cs as the EEL. The regular device(device structure: ITO/GO/P3HT:PCBM/GO-Cs/Al) showed theVOC of 0.61 V, JSC of 10.30 mA cm�2, FF of 0.59, and PCE of3.67%. The inverted device (device structure: ITO/GO-Cs/P3HT:PCBM/GO/Al) showed the VOC of 0.51 V, JSC of 10.69 mAcm�2, FF of 0.54, and PCE of 2.97%. The normal and inverteddevices based on GO hole- and GO-Cs electron-extraction layersboth showed comparable photovoltaic performance to the cor-responding standard BHJ solar cells with the state-of-the-arthole- and electron-extraction layers.
On the other hand, Qu et al.63 have developed a r-GO/fullerene composite as the EEL for PSCs. These authors rst
1304 | Energy Environ. Sci., 2014, 7, 1297–1306
synthesized a fullerene derivative bearing a pyrene “anchor”unit, which was then non-covalently attached to r-GO via the p–p interaction of the pyrene unit and the r-GO carbon basalplane. PSC devices with the r-GO/fullerene composite as the EELexhibited a VOC of 0.64 V, JSC of 9.07 mA cm�2, FF of 0.62, andPCE of 3.89%, which was higher than 3.39% of the controldevice without EEL.
Wang et al.64 have fabricated highly efficient PSCs with aGO/TiOx bilayer as the EEL and poly[N-90 0-heptadecanyl-2,7-carbazole-alt-5,5-(40,70-di-2-thienyl-20,10,30-benzothiadiazole)]
(PCDTBT):[6,6]-phenyl C71 butyric acid methyl ester (PC71BM) asthe active layer. For the device fabrication (Fig. 11a–d), a GO layerwas deposited by graphene stamping transfer from copper foilwith a thermal-release tape, followed by oxidation with HNO3.The PSC device thus fabricated showed a PCE as high as 7.5%,along with a VOC of 0.88 V, JSC of 12.40 mA cm�2, and FF of 0.68.The GO layer was reported to have a proper work function of4.3 eV, close to the LUMO energy level of PC71BM, leading toefficient electron transport. Thus, the GO layer played animportant role in improving the JSC and PCE. Furthermore, thedevice with GO exhibited a much higher stability (3% PCE decay)than the corresponding device without the GO interlayer (56%PCE decay).
4. Concluding remarks
Charge extraction layers play an important role in improving thepower conversion efficiency and lifetime of PSCs. Owing to theirexcellent solution processability, unique two-dimensionalstructure, functionalization-induced work function tunability,and ambipolar transporting ability, GO and its derivatives havequickly emerged as a new class of efficient hole and electronextraction materials in PSCs. We have summarized recentprogress in this newly emerging and exciting research eld. AsHELs, GO materials have showed better performance than thestate-of-the-art HEL (i.e. PEDOT:PSS) and are very competitivewith other novel HELs, including the solution processablemetal oxides. For the EEL application, GO has already showedgreat promise, though it is still a research eld in infancy.
Current research on GO as a charge extraction layer focuseson PSC devices with P3HT:PCBM as the active layer and ITO asthe electrode. However, P3HT:PCBM suffers from low deviceefficiency while many highly efficient donor or acceptor mate-rials have already been developed.65,66 ITO will not be the ulti-mate choice for transparent electrodes of exible PSCs becauseof its high cost and brittleness. Various transparent electrodesbased on conducting polymers, metal nanowires, carbonnanotubes, and graphene have emerged as the exible electrodeto replace ITO.67,68 Therefore, novel GO-based charge extractionlayers should be developed to match the newly developed effi-cient donor and acceptor materials in the active layer and thenew transparent electrodes. For instance, new efficient donormaterials always have lower-lying HOMO level than that ofP3HT. For GO to form Ohmic contact with these donor mate-rials to facilitate hole extraction, the work function of GO needsto be lowered. In contrast to the hydrophilic ITO electrode,graphene transparent electrodes are oen hydrophobic withwhich the deposition of a thin lm of GO as the chargeextraction layer by solution processing is difficult, if notimpossible. Hence, hydrophobic GO charge extraction layersneed to be developed to match graphene electrodes. Thecombination of GO and its derivatives with other chargeextraction materials will likely afford superior PSC deviceperformance. Continued research efforts in this emerging eldcould give birth to a ourishing area of photovoltaictechnologies.
LD is grateful for the nancial support from AFOSR (FA9550-12-1-0069). JL thanks the nancial support by the Nature ScienceFoundation of China (no. 51373165), the 973 project (no.2014CB643504), “Youth Thousand Talents Program” of China.
Notes and references
1 G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger,Science, 1995, 270, 1789–1791.
2 B. C. Thompson and J. M. J. Frechet, Angew. Chem., Int. Ed.,2007, 47, 58–77.
3 Y. Y. Liang, Z. Xu, J. B. Xia, S. T. Tsai, Y. Wu, G. Li, C. Ray andL. P. Yu, Adv. Mater., 2010, 22, E135–E138.
4 H. Y. Chen, J. H. Hou, S. Q. Zhang, Y. Y. Liang, G. W. Yang,Y. Yang, L. P. Yu, Y. Wu and G. Li, Nat. Photonics, 2009, 3,649–653.
5 K. H. Hendriks, G. H. L. Heintges, V. S. Gevaerts,M. M. Wienk and R. A. J. Janssen, Angew. Chem., Int. Ed.,2013, 52, 8341–8344.
6 B. Azzopardi, C. J. M. Emmott, A. Urbina, F. C. Krebs,J. Mutale and J. Nelson, Energy Environ. Sci., 2011, 4, 3741–3753.
7 F. C. Krebs, T. D. Nielsen, J. Fyenbo, M. Wadstrøm andM. S. Pedersen, Energy Environ. Sci., 2010, 3, 512–525.
8 G. Li, R. Zhu and Y. Yang, Nat. Photonics, 2012, 6, 153–161.9 H. J. Park, H. Kim, J. Y. Lee, T. Lee and L. J. Guo, EnergyEnviron. Sci., 2013, 6, 2203–2210.
10 Z. C. He, C. M. Zhong, S. J. Su, M. Xu, H. B. Wu and Y. Cao,Nat. Photonics, 2012, 6, 591–595.
11 J. B. You, L. T. Dou, K. Yoshimura, T. Kato, K. Ohya,T. Moriarty, K. Emery, C. C. Chen, J. Gao, G. Li andY. Yang, Nat. Commun., 2013, 4, 1446–1455.
12 L. S. Hung and C. H. Chen,Mater. Sci. Eng., R, 2002, 39, 143–222.
13 H. Park, S. Chang, M. Smith, S. Gradecak and J. Kong, Sci.Rep., 2013, 3, 1581.
14 R. Steim, F. R. Kogler and C. J. Brabec, J. Mater. Chem., 2010,20, 2499–2512.
15 H. L. Yip and A. K. Y. Jen, Energy Environ. Sci., 2012, 5, 5994–6011.
16 L. M. Chen, Z. Xu, Z. R. Hong and Y. Yang, J. Mater. Chem.,2010, 20, 2575–2598.
17 R. Po, C. Carbonera, A. Bernardi and N. Camaioni, EnergyEnviron. Sci., 2011, 4, 285–310.
18 M. Girtan andM. Rusu, Sol. Energy Mater. Sol. Cells, 2010, 94,446–450.
19 M. Jorgensenm, K. Norrrman and F. C. Krebs, Sol. EnergyMater. Sol. Cells, 2008, 92, 686–714.
20 S. Chen, J. R. Manders, S. W. Tsang and F. So, J. Mater.Chem., 2012, 22, 24202–24212.
21 J. B. You, C. C. Chen, L. Dou, S. Murase, H. S. Duan,S. A. Hawks, T. Xu, H. J. Son, L. P. Yu, G. Li and Y. Yang,Adv. Mater., 2012, 24, 5267–5272.
22 J. J. Jasieniak, J. Seier, J. Jo, T. Mates and A. J. Heeger, Adv.Funct. Mater., 2012, 22, 2594–2605.
23 A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.24 M. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2010, 110,
132–145.25 D. Chen, H. Zhang, Y. Liu and J. H. Li, Energy Environ. Sci.,
2013, 6, 1362–1387.26 Y. H. Ng, S. Ikeda, M. Matsumura and R. Amal, Energy
Environ. Sci., 2012, 5, 9307–9318.27 W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958,
80, 1339.28 S. Park and R. S. Ruoff, Nat. Nanotechnol., 2009, 4, 217–224.29 G. Eda and M. Chhowalla, Adv. Mater., 2010, 22, 2392–2415.30 Y. Zhu, S. Murali, W. Cai, X. Li, J. W. Suk, J. R. Potts and
R. S. Ruoff, Adv. Mater., 2010, 22, 3906–3924.31 L. M. Dai, D. W. Chang, J. B. Baek and W. Lu, Small, 2012, 8,
1130–1166.32 J. Liu, Y. H. Xue, M. Zhang and L. M. Dai,MRS Bull., 2012, 37,
1265–1272.33 L. M. Dai, Acc. Chem. Res., 2013, 46, 31–42.34 X. J. Wan, G. K. Long, L. Huang and Y. S. Chen, Adv. Mater.,
2011, 23, 5342–5348.35 A. Iwan and A. Chuchmala, Prog. Polym. Sci., 2012, 37, 1805–
1828.36 Y. Wang, S. W. Tong, X. F. Xu, B. Ozyilmaz and K. P. Loh, Adv.
Mater., 2011, 23, 1514–1518.37 Y. Li, Y. Hu, Y. Zhao, G. Q. Shi, L. E. Deng, Y. B. Hou and
L. T. Qu, Adv. Mater., 2011, 23, 776–780.38 S. S. Li, K. H. Tu, C. C. Lin, C. W. Chen and M. Chhowalla,
ACS Nano, 2010, 4, 3169–3174.39 Y. Gao, H. L. Yip, S. K. Hau, K. M. Malley, N. C. Cho,
H. Z. Chen and A. K. Y. Jen, Appl. Phys. Lett., 2010, 97,203306.
40 Y. Gao, H. L. Yip, K. S. Chen, K. O'Malle, M. O. Acton, Y. Sun,G. Ting, H. Z. Chen and A. K. Y. Jen, Adv. Mater., 2011, 23,1903–1908.
41 D. R. Dreyer, S. Park, C. W. Bielawski and R. S. Ruoff, Chem.Soc. Rev., 2010, 39, 228–240.
42 S. Mao, H. H. Pu and J. H. Chen, RSC Adv., 2012, 2, 2643–2662.
43 S. F. Pei and H.-M. Cheng, Carbon, 2012, 50, 3210–3228.44 J. M. Yun, J. S. Yeo, J. Kim, H. G. Jeong, D. Y. Kim, Y. J. Noh,
S. S. Kim, B. C. Ku and S. I. Na, Adv. Mater., 2011, 23, 4923–4928.
45 J. Liu, Y. H. Xue and L. M. Dai, J. Phys. Chem. Lett., 2012, 3,1928–1933.
46 H. P. Kim, A. R. B. M. Yusoff and J. Jang, Sol. Energy Mater.Sol. Cells, 2013, 110, 87–93.
47 Y. J. Joen, J. M. Yun, D. Y. Kim, S. I. Na and S. S. Kim, Sol.Energy Mater. Sol. Cells, 2012, 105, 96–102.
48 X. D. Liu, H. Kim and L. J. Guo, Org. Electron., 2013, 14, 591–598.
1306 | Energy Environ. Sci., 2014, 7, 1297–1306
49 J. Kim, V. C. Tung and J. X. Huang, Adv. Energy Mater., 2011,1, 1052–1057.
50 I. P. Murray, S. J. Lou, L. J. Cote, S. Loser, C. J. Kadleck, T. Xu,J. M. Szarko, B. S. Rolczynski, J. E. Johns, J. X. Huang,L. P. Yu, L. X. Chen, T. J. Marks and M. C. Hersam, J. Phys.Chem. Lett., 2011, 2, 3006–3012.
51 D. Yang, L. Y. Zhou, L. C. Chen, B. Zhao, J. Zhang and C. Li,Chem. Commun., 2012, 48, 8078–8080.
52 V. C. Tung, J. Kim and J. X. Huang, Adv. Energy Mater., 2012,2, 299–303.
53 B. Yin, Q. Liu, L. Y. Yang, X. M. Wu, Z. F. Liu, Y. L. Hua,S. G. Yin and Y. S. Chen, J. Nanosci. Nanotechnol., 2010, 10,1934–1938.
54 G. Q. Fan, Q. Q. Zhuo, J. J. Zhu, Z. Q. Xu, P. P. Cheng, Y. Q. Li,X. H. Sun, S. T. Leeb and J. X. Tang, J. Mater. Chem., 2012, 22,15614–15619.
55 V. C. Tung, J. Kim, L. J. Cote and J. X. Huang, J. Am. Chem.Soc., 2011, 133, 9262–9265.
56 M. S. Ryu and J. Jang, Sol. Energy Mater. Sol. Cells, 2011, 95,2893–2896.
57 Y. H. Chao, J. S. Wu, C. E. Wu, J. F. Jheng, C. L. Wang andC. S. Hsu, Adv. Energy Mater., 2013, 3, 1279–1285.
58 Y. Park, K. S. Choi and S. Y. Kim, Phys. Status Solidi A, 2012,209, 1363–1368.
59 K. S. Choi, Y. Park and S. Y. Kim, J. Nanosci. Nanotechnol.,2013, 13, 3282–3287.
60 J. Liu, G.-H. Kim, Y. H. Xue, J. Y. Kim, J.-B. Baek, M. Durstockand L. M. Dai, Adv. Mater., 2013, DOI: 10.1002/adma.201302987.
61 S. Wang, P. K. Ang, Z. Q. Wang, A. L. L. Tang,J. T. L. Thompson and K. P. Loh, Nano Lett., 2010, 10,92–98.
62 J. Liu, Y. H. Xue, Y. X. Gao, D. S. Yu, M. Durstock andL. M. Dai, Adv. Mater., 2012, 24, 2228–2233.
63 S. X. Qu, M. H. Li, L. X. Xie, X. Huang, J. G. Yang, N. Wangand S. F. Yang, ACS Nano, 2013, 7, 4070–4081.
64 D. H. Wang, J. K. Kim, J. H. Seo, I. Park, B. H. Hong,J. H. Park and A. J. Heeger, Angew. Chem., Int. Ed., 2013,52, 2874–2880.
65 J. Liu, H. Choi, J. Y. Kim, C. Bailey, M. Durstock and L. Dai,Adv. Mater., 2012, 24, 538–542.
66 T. B. Yang, M. Wang, C. H. Duan, X. W. Hu, L. Huang,J. B. Peng, F. Huang and X. Gong, Energy Environ. Sci.,2012, 5, 8208–8214.
67 S. Bae, H. Kim, Y. Lee, X. F. Xu, J.-S. Park, Y. Zheng,J. Balakrishnan, Y. Lei, H. R. Kim, Y. I. Song, Y. J. Kim,B. Oyilmaz, J. H. Ahn, B. H. Hong and S. Lijima, Nat.Nanotechnol., 2010, 5, 574–578.
68 D. S. Hecht, L. B. Hu and G. Irvin, Adv. Mater., 2011, 23,1482–1513.
![A Decade of Graphene Research - Production, Apps & Outlook [Randviir, 2014]](https://static.documents.pub/doc/80x56/55cf91de550346f57b915639/a-decade-of-graphene-research-production-apps-outlook-randviir-2014.jpg)
